Implementation of the QoS framework using fog computing to predict COVID-19 disease at early stage

Abstract

Purpose The purpose of this paper is to provide more accurate structure that allows the estimation of coronavirus (COVID-19) at a very early stage with ultra-low latency. The machine learning algorithms are used to evaluate the past medical details of the patients and forecast COVID-19 positive cases, which can aid in lowering costs and distinctively enhance the standard of treatment at hospitals. Design/methodology/approach In this paper, artificial intelligence (AI) and cloud/fog computing are integrated to strengthen COVID-19 patient prediction. A delay-sensitive efficient framework for the prediction of COVID-19 at an early stage is proposed. A novel similarity measure-based random forest classifier is proposed to increase the efficiency of the framework. Findings The performance of the framework is checked with various quality of service parameters such as delay, network usage, RAM usages and energy consumption, whereas classification accuracy, recall, precision, kappa static and root mean square error is used for the proposed classifier. Results show the effectiveness of the proposed framework. Originality/value AI and cloud/fog computing are integrated to strengthen COVID-19 patient prediction. A novel similarity measure-based random forest classifier with more than 80% accuracy is proposed to increase the efficiency of the framework.